Cyclodextrin assisted enantiomeric recognition of benzo[de]isoquinoline-1, 3-dione derived amino acids

Article abstract of DOI:10.1016/j.tet.2004.11.021

Spectroscopic evidence for enantiomeric recognition of properly modified amino acids through the cyclodextrin assisted formation of polymer like self-assemblies is presented. The requirements for the formation of these assemblies through aromatic π-π stac

Full text of DOI:10.1016/j.tet.2004.11.021

Tetrahedron 61 (2005) 919–926
Cyclodextrin assisted enantiomeric recognition of
benzo[de]isoquinoline-1,3-dione derived amino acids
*
Branko S. Jursic and Paresh K. Patel
Department of Chemistry, University of New Orleans, New Orleans, LA 70148 USA
Received 20 July 2004; revised 3 November 2004; accepted 5 November 2004
Available online 30 November 2004
Abstract—Spectroscopic evidence for enantiomeric recognition of properly modified amino acids through the cyclodextrin assisted
formation of polymer like self-assemblies is presented. The requirements for the formation of these assemblies through aromatic p–p
stacking are discussed. It is suggested that this approach can be used for enantiomeric separation of chiral compounds that contain both
electron-rich and electron-poor moieties.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
can form diasteriomeric inclusion complexes with racemic
aromatic compounds. Based on the knowledge that
diasteriomers have different physical properties, it is
reasonable to expect that one of the diasteriomeric inclusion
complexes will crystallize out from the racemic aqueous
cyclodextrin solution more readily.
It is now well accepted that marketing only one enantiomer
of a chiral drug is crucial for the pharmaceutical industry.1
There are two commonly used approaches for preparing the
enantiomerically pure compounds needed for chiral drug
marketing. These approaches include; (a) development of
preparation procedures that yield only the desirable
enantiomer2 and (b) using preparation procedures that
generate both enantiomers and then purifying the racemic
mixture using expensive chiral separation techniques. The
first approach is very expensive because it is necessary to
use either chiral intermediates or chiral catalysts to obtain
the desired enantiomerically pure product, while the second
approach, which results in the production of racemates,
causes problems with waste management and chiral
separations. Therefore, it would be ideal to develop new
chemistry that can address both of these problems by
applying inexpensive preparation procedures for the
production of racemic products, which can then be
transferred from a racemic mixture to the desired enantio-
merically pure targeted compound.
It has also been well established that stronger enantiomeric
recognition can be accomplished if two enantiomers
compete to bind to a multi-chiral resolving molecule.4 The
same effect should be present if two enantiomers are
competing to bind to a homochiral self-assembly polymer
made from one of the enantiomers. There are many
nonbonding interactions (such as hydrogen bonding,
electrostatic interactions, p–p-aromatic stacking. dipole–
dipole, etc.) responsible for the formation of molecular
associates.5 Aromatic p–p stacking interactions are crucial
for formation of molecular associates, interactions between
substrate and enzymes, biomolecular associates, organic
molecule conformation, etc.6 Furthermore, multi-chirality
and multi-point interactions are very important for enantio-
meric recognition.7
There are many drugs that have aromatic moieties and are
currently marketed as racemic mixtures. These drugs
include Prozac, Escitalopram, and Clopidogrel to name a
few.1d It is also known that cyclodextrins are capable of
forming stable complexes with aromatic compounds.3
Therefore, it is reasonable to propose that cyclodextrins
The ideal cyclodextrin assisted molecular associate, which
has all of these requirements, is presented in Figure 1.
Molecule M has four moieties that should be essential for
the formation of homochiral polymer-like molecular
associates. The two aromatic moieties are complementary
to one another, and are the electron-donor and the electron-
acceptor. This aromatic complementarity enables the
molecule to form molecular associates through p–p
stacking. In general, the p–p stacking of aromatic
complexes are very weak, therefore these complexes should
Keywords: Enantiomers; Enantiomeric recognition; Cyclodextrin
complexation.
* Corresponding author. Tel.: C1 504 280 7090; fax: C1 504 280 6860;
e-mail: bsjcm@uno.edu
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.11.021

920
B. S. Jursic, P. K. Patel / Tetrahedron 61 (2005) 919–926
Figure 1. Possible pathway to cyclodextrin assisted formation of homochiral polymer-like associate.
be stabilized through their binding into the cyclodextrin
cavity (Fig. 1). During the formation of this cyclodextrin
assisted molecular associate, two aromatic moieties, each
from different M molecules, are now ready to form a new
cyclodextrin inclusion complex and molecular associate
with a new M molecule. In this way, the polymer-like
molecular associate would be formed.
electronic properties (Fig. 2). The AM111 computed frontier
molecular orbital (HOMO–LUMO) energies suggest that
toluene is electronically neutral, indole is electron rich and,
benzoisoquinoline-1,3-dione is electron poor (Fig. 2). If
these molecular moieties are incorporated into amino acids,
then the requirement for molecule
accomplished.
M has been
The presence of polar groups, such as carboxylates, enable
this polymer-like associate to interact with water surround-
ings. On the other hand, chirality of M makes the molecular
associate diasteriomeric and stereoselective for incorporat-
ing only one enantiomer of M into the homochiral
molecular associate.
Three amino acids were selected to test the validity of this
hypothesis: alanine (without aromatic moiety), phenyl-
alanine (neutral aromatic electronic properties), and
trypthophan (electron-rich electronic properties). Through
amino group replacement with benzo[de]isoquinoline-1,3-
dione, new molecular systems 1a, 1b, and 1c were
generated. As expected, the LUMO orbital energy for all
these three molecules is similar (because each has the same
electron poor aromatic moiety) while the HOMO energies
vary dramatically (Fig. 3) with the expected best comple-
mentarily of HOMO–LUMO interactions with 1c. The AM1
estimated LUMO–HOMO energy difference as measured
for p–p aromatic stacking capabilities are 8.00, 7.85, and
7.69 eV for 1a, 1b, and 1c, respectively (the lower LUMO–
HOMO energy gap the higher capability to form molecular
associates through p–p molecular stacking8).
To test the validity of the proposed cyclodextrin assisted
self-assembly approach to enantiomeric recognition, the
proper molecular system must be selected. Amino acids
seem to be the best molecular choice for this study, provided
that the amino acids have both electron-rich and electron-
poor aromatic moieties. None of the natural amino acids
satisfy this requirement; therefore, through simple chemical
modifications of natural amino acids this requirement can be
achieved. For instance, simple aromatic molecules such as
methylindole, toluene, and methylbenzoisoquinoline-1,3-
dione (Fig. 2) have a noticeable difference in aromatic
The synthetic conditions for the preparation of these amino
Figure 2. The AM1 estimated HOMO–LUMO energies for the aromatic moieties.
Figure 3. The AM1 estimated HOMO–LUMO energies for three modified amino acids.

B. S. Jursic, P. K. Patel / Tetrahedron 61 (2005) 919–926
921
Scheme 1. Preparation of electron-deficient alanine, phenylalanine, and tryptophan.
acid derivatives are presented in Scheme 1. Reactions are
performed with the amino acid and benzo[de]isochromene-
1,3-dione in DMSO for alanine and pyridine for phenyl-
alanine and tryptophan (Scheme 1). The isolated yields
range between 80 and 95%.
enantiomeric ratio show spectroscopic recognition. On the
other hand, there is NMR enantiomeric recognition of non-
racemic 1c (Fig. 4). The spectroscopic recognition is not due
to the formation of molecular associates through hydrogen
bonding, nor through electrostatic interactions between the
carboxylate group and sodium, because these interactions
are also present in 1a and 1b. The only reasonable
explanation is that molecule 1c can form molecular
associates through p–p stacking between the indole and
benzo[de]isochromene-1,3-dione moieties.10
To demonstrate the existence of weak nonbonding inter-
actions, as well as enantiomeric recognition when one of the
enantiomers acts as resolving agent,9 spectroscopic studies
of racemic and non-racemic highly concentrated solutions
of 1a, 1b, and 1c were studied. These compounds are
soluble in DMSO, therefore this was our solvent of choice.
Both concentration (0.0001–0.3 M) and molar ratio (1:10,
1:5, 1:1, 5:1, and 10:1) of the S and R enantiomers were
varied. In none of our NMR studies with all three
compounds were we able to observe any NMR evidence
for enantiomeric recognition.
If this nonbonding p–p stacking is responsible for
enantiomeric recognition, then the presence of cyclodextrin
will enhance NMR spectroscopic recognition. This is due to
the fact that both aromatic moieties tend to bind into the
cyclodextrin cavity, and then orient themselves in the
proper way for the aromatic stacking and for strong
homochiral molecular associates. This is perfectly demon-
strated in Figure 5 for g-cyclodextrin assisted enantiomeric
recognition of the racemic 1c. The best results are obtained
in g-cyclodextrin in comparison with b-cyclodextrin, while
no enantiomeric recognition was observed in a-cyclo-
dextrin. This finding is not surprising considering that the
cyclodextrin cavity must be large enough to accommodate
two aromatic rings.
For the study of the cyclodextrin assisted molecular
complexes, water is a much better solvent of choice.
However, none of the studied compounds is sufficiently
soluble in water to perform reliable NMR studies. There-
fore, sodium salts of 1a, 1b, and 1c were used for the NMR
spectroscopic studies. Neither 1a nor 1b non-racemic
mixtures at any concentration range up to 0.1 M and any
Figure 4. The NMR spectra of aromatic portion of 1c (0.1 M) in aqueous NaHCO3 (0.3 M) with different enantiomer ratios.
Figure 5. A portion of NMR spectra of racemic 1c (0.001 M) in aqueous NaHCO3 (0.003 M) with a, b, and g-CD (0.01 M), respectively.

922
B. S. Jursic, P. K. Patel / Tetrahedron 61 (2005) 919–926
Figure 6. A portion of NMR spectra of 1c (0.001 M) in aqueous NaHCO3 (0.003 M) and g-CD (0.01 M).
There are two sets of peaks in racemate NMR in presence of
g-CD. It is also possible to easily recognize each enantiomer
peak if individual enantiomers NMR in g-CD compared
with the racemate NMR in g-CD (Fig. 6).
clearly shows the formation of 2:1 and 1:2 g-CD complexes
with 1c (signals at 1488 and 1032, Fig. 8). All the other
molecular complexes are present as well.
These are very encouraging results, but there is no
experimental evidence for the formation of polymer-like
cyclodextrin assisted molecular assemblies presented in
Figure 1. For this reason, we performed MALDI MS spectra
analysis of the aqueous g-cyclodextrin solution of 1c. If the
structural patterns presented in Figure 1 are present, then
typical fragmentation similar to the fragmentation of
polymers should be observed in the MALDI MS spectra.
All possible fragments were detected with lower intensity
for larger molecular fragments (Fig. 9).
Enantiomeric recognition can be observed even if there is no
formation of the molecular associates described in Figure 1.
The enantiomer binding into the cyclodextrin cavity can
initiate the chiral environment necessary to observe
spectroscopic differences. In a-cyclodextrin no NMR
spectroscopic recognition was observed due to fact that
neither the indole nor benzo[de]isochromene-1,3-dione
aromatic moieties can bind into the a-cyclodextrin cavity.
The b-cyclodextrin cavity is sufficiently large to accommo-
date the indole ring. If there is formation of molecular
associates higher than the 1:1 cyclodextrin complex with
respect to 1c, we should observe this in electrospray mass
spectroscopy. By performing this experiment it is clearly
demonstrated that the cyclodextrin inclusion complex with
1c was formed (Fig. 7) but no higher order (1:2 or 2:1)
complex between b-CD and 1c was observed.
To better understand the binding of these guest compounds
with host cyclodextrins, association constants (K, MK1
)
were measured by 1H NMR (500 MHz) at different
temperature. In Table 1 are listed association constant K,
standard free energy DG8, standard entropy term TDS8 and
standard enthalpy DH8 of guest 1c with b and g
cyclodextrin. The chemical shift change of NMR signal of
guest 1c in presence of a cyclodextrine was too small to
measure (as shown in Fig. 5). So the association constant is
too small to measure practically. The solution of 1c
Now we turn our attention to the molecular system that
shows the best NMR enantiomeric recognition; g-CD
complexes with 1c (Fig. 5). The electrospray mass spectra
Figure 7. Negative electrospray mass spectrophotometry of 1cR (0.001 mol) in aqueous NaHCO3 (0.003 M) and b-cyclodextrin (b-CD, 0.01 M).
Figure 8. Negative electrospray mass spectra of 1cS (0.001 M) in aqueous NaHCO3 (3!10K3 M) and g-cyclodextrin (g-CD, 10K2 M).

B. S. Jursic, P. K. Patel / Tetrahedron 61 (2005) 919–926
923
Figure 9. Positive MALDI-MS spectra of aqueous g-CD and 1c.
Table 1. Optical rotations of guest compounds 1a, 1b and 1c
more than seven times higher for g-CD compare to the
association constant for b-CD. These results are consistent
with what we understand from Figure 5. The cyclodextrin
cavity must be large enough to accommodate two aromatic
rings which g-CD have. Also the association constant for
L-enantiomer is higher than that of D-enantiomer.
Guest
compound
Optical rotation
L-enantiomer
D-enantiomer
1a
1b
1c
K19.5 (c 0.1, MeOH)
K31.6 (c 1, MeOH)
K17.0 (c 0.1, H2O)
C18.0 (c 0.1, MeOH)
C28.4 (c 1, MeOH)
C16.2 (c 0.1, H2O)
To show that the presence of the electron-poor aromatic
moiety is crucial for the formation of the molecular
assemblies discussed above, as well as make the amino
acid derivative very soluble in water, mono amino acid
amides with succinic acid were prepared (Scheme 2). Each
and every of these reactions requires specific conditions,
which were optimized by monitoring the reaction by NMR
spectroscopic studies. After all reaction conditions were
optimized, the isolated yields were higher than 85%
(Table 3).
(0.001 M) in aqueous NaHCO3 (3!10K3 M) was titrated
with host solution at different temperature. Each time the
change in chemical shift was measured. Non-linear
regression analysis using origin 6.1 (Aston Scientific Ltd)
was used to generate the association constant Ka according
to the equation12:
DmaxKa½Hꢀ
D Z
1 CKa½Hꢀ
All of these compounds are very soluble in water, which
makes it possible to perform the NMR spectroscopic studies
in DMSO, CDCl3, and water. We did not observe NMR
enantiomeric recognition in DMSO and water, as well as in
aqueous cyclodextrins. We were able to detect through
electrospray MS spectroscopic study the formation of b-CD
inclusion complexes with both 2b and 2c. It seems obvious
that these cyclodextrin inclusion complexes cannot produce
the different NMR environment to cause the NMR spectro-
scopic recognition. When two aromatic rings are together in
the g-CD cavity, as is the case with 1c, the currency of one
where D, the peak shift in ppm, Dmax, the maximum peak
shift in ppm, and [H], the concentration of host. At least two
experiments were performed for each system. Standard free
energy was calculated using DG8ZKRT ln Ka equation.
Standard enthalpy DH8 was determined according to the
Van’t Hoff equation: d ln K/d(1/T)ZKDH8/R where RZ
8.3144 J KK1 molK1 and DH8 in J molK1. TDS8 was
calculated using DG8ZDH8KTDS8 equation.
As shown in Table 2, the association constant of guest 1c is
1
Table 2. H NMR derived thermodynamic parameters for the binding of guest molecule 1c to host b and g-CD
Temperature (K)
L-enantiomer
g-CD
D-enantiomer
b-CD
b-CD
g-CD
298
313
343
KZ63.02G19.63
DG8ZK10.27G0.7
TDS8ZK4.84G2.9
KZ42.57G10.34
DG8ZK9.22G0.6
TDS8ZK5.76G3.1
KZ26.94G3.39
KZ413.03G21.45
DG8ZK14.93G0.13
TDS8ZK9.7G0.12
KZ262.26G11.95
DG8ZK13.81G0.11
TDS8ZK10.83G0.11
KZ113.35G4.86
KZ47.26G6.72
KZ320.93G19.5
DG8ZK9.56G0.3
TDS8ZK28.89G1.5
KZ19.88G3.61
DG8ZK7.41G0.4
TDS8ZK31.05G1.6
KZ6.12G1.26
DG8ZK14.31G0.15
TDS8 K18.71G3.5
KZ210.7G5.73
DG8ZK13.26G0.07
TDS8ZK19.76G3.4
KZ59.26G12.93
DG8ZK8.16G0.3
TDS8ZK6.84G3.4
DH8ZK15G3.7
DG8ZK11.73G0.11
TDS8ZK12.9G0.1
DH8ZK24.63G0.01
DG8ZK4.49G0.5
TDS8ZK33.95G1.7
DH8ZK38.4G1.2
DG8ZK10.12G0.49
TDS8ZK22.9G3.8
DH8ZK33G3.4
All K, DG8, TDS8, DH8 values are in kJ/mol unit.

924
B. S. Jursic, P. K. Patel / Tetrahedron 61 (2005) 919–926
Scheme 2. Preparation of acid-amides of alanine, phenylalanine, and tryptophan.
Table 3. Optical rotations of guest compounds 2a, 2b, and 2c
(1.98 g; 0.01 mol) and alanine (0.89 g; 0.01 mol) was
refluxed for 1 h. Reaction mixture was cooled to room
temperature and slowly added into stirring water (200 mL).
Formed precipitate was separated by filtration, washed
with water (3!30 mL) and dried at 110 8C for a few
hours to afford 2.15 g (80%) pure product. Mp 257–261 8C.
1H NMR (DMSO-d6, 500 MHz): d ppm: 8.51 (dd, 4H,
J1Z10.5 Hz; J2Z7.5 Hz), 7.89 (t, 2H, JZ7.5 Hz), 5.54
(q, 1H, JZ7 Hz), and 1.53 (d, 3H, JZ6.5 Hz). 13C
NMR (DMSO-d6, 500 MHz) d ppm: 171.4, 162.9
(carbonyls), 134.7, 131.3, 131.1, 127.4, 121.7 (aromatic
carbons), 48.5 (chiral carbon CH), 14.5 (methylene carbon).
MS-ESC (CH3OH) m/z 292.2 (65%, MCNaC), 324.1
(85%, MCCH3OHCNaC), 561.3 (100%. 2MCNaC), and
829.8 (50%, 3MCNaC). Anal. Calcd for C15H11NO4
(269.07): C, 66.91; H, 4.12; N, 5.20 Found: C, 66.85; H,
4.21; N, 5.11.
Guest
compound
Optical rotation
L-enantiomer
D-enantiomer
2a
2b
2c
C22.0 (c 1, MeOH)
K14.9 (c 1, MeOH)
C37.1 (c 1, MeOH)
K21.6 (c 1, MeOH)
C13.4 (c 1, MeOH)
K32.8 (c 1, MeOH)
ring causes a change of magnetic properties on the other
ring, which is easily detected through NMR spectroscopic
studies, resulting in spectroscopic recognition of two
enantiomers.
2. Conclusion
In conclusion we can state that there are spectroscopic
evidences for the formation of chiral cyclodextrin assisted
polymer-like aggregates with g-cyclodextrin that accom-
modate two aromatic rings and modified chiral amino acids.
For successful cyclodextrin assistance in the formation of
these kinds of aggregates, modified amino acids must
contain both electron-poor and electron-rich aromatic
moieties. Cyclodextrin with a sufficiently large cavity,
such as g-cyclodextrin must be used to be able to bind both
aromatic rings involved in aromatic p–p stacking.
3.1.2. Preparation of (R)-2-(1,3-dioxo-1H,3H-benzo[de]-
isoquinolin-2-yl)propionic acid (1aR). The stereoisomer R
was prepared in 87% yield by following procedure for
preparation of S isomer. The NMR and MS-ES spectra of R
and S stereoisomers are identical.
3.1.3. Preparation of (S)-2-(1,3-dioxo-1H,3H-benzo[de]-
isoquinolin-2-yl)-3-phenyl-propionic acid (1bS). Pyridine
solution (150 mL) of benzo[de]isochromene-1,3-dione
(0.99 g; 5 mmol) and L-phenylalanine (0.825 g; 5 mmol)
was refluxed for 6 h. Volume of the reaction mixture was
reduced to w5 mL and hot reaction mixture was added to
ice cooled aqueous hydrochloric acid (150 mL water and 50
concd HCl). Formed solid precipitate was separated by
filtration, washed with water (3!20 mL) and dried at
110 8C for a few hours to afford 1.6 g (93%) pure product.
3. Experimental
3.1. General
Melting points were taken on an Electrothermal IA 9000
Digital Melting Point Apparatus and are uncorrected. The
1H and 13C NMR spectra were run on Varian Unity 400 and
Varian INOVA 500 MHz spectrophotometer with DMSO-
d6 as a solvent and internal standard (2.50 and 35.91 ppm for
1H and 13C NMR, respectively). The mass spectra were
recorded on a Micromass Quattro 2 Triple Quadropole Mass
Spectrometer; Optical rotations of guest compounds were
detected at 25 8C with the light of sodium D-line (589 nm)
using Autopo III automatic polarimeter Rudolph Research,
Flanders, New Jersy. Elemental Analysis was performed by
Atlantic Microlab, Inc., Norcross, GA.
1
Mp 250–252 8C H NMR (DMSO-d6, 500 MHz) d ppm:
8.41 (d, 2H naphthalene ring, JZ7 Hz), 8.37 (d, 2H
naphthalene ring, JZ7 Hz), 7.78 (t, 2H naphthalene ring,
JZ7.5 Hz,), 7.14 (d, 2H, JZ7.5 Hz), 7.07 (2H, t, JZ
7.5 Hz) 7.00 (t, 1H, JZ7.5 Hz) 5.92 (dd,1H, J1Z10 Hz;
J2Z5.5 Hz) 3.58 (dd, 1H, J1Z14 Hz; J2Z5.5 Hz), and 3.39
(dd,1H, J1Z14.5 Hz; J2Z10 Hz). 13C NMR (DMSO-d6,
500 MHz) d ppm: 171.8, 163.0 (carbonyls), 137.9, 134.7,
131.2, 129.0, 128.1, 127.5, 127.3, 127.2, 126.3, 121.2 (10
aromatic carbons), 53.9 (chiral carbon CH), and 24.3
(methylene carbon). MS-ESK m/z 344 (100% MKHC).
Anal. Calcd for C21H15NO4 (345.10): C, 70.03; H, 4.38; N,
4.06. Found: C, 69.91; H, 4.42; N, 3.95.
3.1.1. Preparation of (S)-2-(1,3-dioxo-1H,3H-benzo[de]-
isoquinolin-2-yl)propionic acid (1aS). Dimethyl sulfoxide
(30 mL) solution of benzo[de]isochromene-1,3-dione

B. S. Jursic, P. K. Patel / Tetrahedron 61 (2005) 919–926
925
3.1.4. Preparation of (R)-2-(1,3-dioxo-1H,3H-benzo[de]-
isoquinolin-2-yl)-3-phenyl-propionic acid (1bR). Stereo-
isomer R was prepared in 91% yield and has same
spectroscopic characteristics as stereoisomer S.
suspension of phenylalanine (1.65 g; 0.01 mol) and succinic
anhydride (1.0 g; 0.01 mol) was refluxed until the suspen-
sion becomes clear solution (approximately 30 h). Solvent
was evaporated under reduced pressure. Solid residue was
scurried in petroleum ether (100 mL), separated by
filtration, and washed with petroleum ether (3!20 mL)
and dried at 60 8C for a few hours. The isolated yield is 2.5 g
(95%) of pure product. Mp 127–129 8C. 1H NMR (DMSO-
d6, 500 MHz) d ppm: 8.15 (d, 1H, amide NH, JZ8 Hz) 7.25
(m, 5H benzene ring), 4.50 (ddd, 1H chiral), 3.05 (dd, 1H,
J1Z13.5 Hz; J2Z5 Hz), 2.86 (dd, 1H, J1Z13.5 Hz; J2Z
9.5 Hz), and 2.35 (s, 4H aliphatic two CH2 group). 13C
NMR (DMSO-d6, 500 MHz) d ppm: 73.8, 173.1, 171.1
(carbonyls), 129.2, 128.2, 126.4 (aromatic carbons), 53.6,
37.0, 29.9, and 29.1 (four aliphatic carbon). MS-ESC m/z
266.1 (22%, MCHC) and 288.1 (100%, MCNaC). Anal.
Calcd for C13H15NO5 (265.10): C, 58.86; H, 5.70; N, 5.28
Found: C, 58.92; H, 5.77; N, 5.15.
3.1.5. Preparation of (S)-2-(1,3-dioxo-1H,3H-benzo[de]-
isoquinolin-2-yl)-3-(1H-indol-3-yl)-propionic acid (1cS).
Dimethyl sulfoxide (30 mL) solution of benzo[de]iso-
chromene-1,3-dione (0.99 g; 5 mmol) and L-tryptophan
(1.04 g; 5 mmol) was heated at 150 8C for 30 min. Red
colored reaction mixture was slowly added into stirring
water (200 mL). Formed yellow precipitate was separated
by filtration, washed with water (3!20 mL) and dried at
110 8C to give 1.6 g (83%) of pure product. Mp 235–239 8C.
1H NMR (DMSO-d6, 400 MHz) d ppm: 12.80 (br s, 1H),
10.63 (1H, s), 8.44 (dd, 4H, J1Z4.4 Hz, J2Z7.2 Hz), 7.85
(t, 2H, JZ7.6 Hz), 7.46 (d, 1H, JZ8.0 Hz), 7.17 (d, 1H, JZ
7.6 Hz), 6.99 (s, 1H,), 6.92 (t, 1H, JZ7.6 Hz), 6.79 (t, 1H,
JZ7.2 Hz), 5.88 (dd, 1H, J1Z9.6 Hz, J2Z5.6 Hz), 3.65
(dd, 1H, J1Z14.8 Hz, J2Z5.6 Hz), 5.54 (dd, 1H, J1Z
14.8 Hz, J2Z9.6 Hz), 13C NMR (DMSO-d6, 500 MHz) d
ppm: 171.0, 163.1, 135.9, 134.7, 131.2, 127.4, 127.2, 123.5,
121.4, 120.8, 118.2, 118.0, 111.3, 110.3, 53.7, and
24.0 ppm. ESIC (CH3CO2H) m/z 385 (100, MCHC) and
769 (45%, 2MCHC). Anal. Calcd for C23H16N2O4
(384.38): C, 71.87; H, 4.20; N, 7.29 Found: C, 71.55; H,
4.31; N, 7.18.
3.1.10. Preparation of (R)-N-(1-carboxy-2-phenylethyl)-
succinamic acid (2bR). The R stereoisomer was prepared in
93% isolated yield by following the procedure described for
the S stereoisomer. The spectral characteristics for this
compound are identical to the spectral characteristics of the
S stereoisomer.
3.1.11. Preparation of (S)-N-[1-carboxy-2-(1H-indol-3-
yl)ethyl]succinamic acid (2cS). Tetrahydrofuran (1 L)
suspension of tryptophan (2 g; 0.01 mol) and succinic
anhydride (1 g; 0.01 mol) was refluxed until it became
solution (approximately 40 h). Solvent was evaporated and
solid residue was mixed with petroleum ether (200 mL).
Solid material was separated by filtration, washed with
petroleum ether (3!20 mL), and dried at 60 8C for several
hours to afford 2.8 g (92%) of pure product. Mp (hydro-
3.1.6. Preparation of (R)-2-(1,3-dioxo-1H,3H-benzo[de]-
isoquinolin-2-yl)-3-(1H-indol-3-yl)-propionic acid (1cR).
The R stereoisomer has same spectroscopic characteristics
as S isomer and is prepared in 87% isolated yield by
following the S preparation procedure.
3.1.7. Preparation of (S)-N-(1-carboxyethyl)succinamic
acid (2aS). Water solution (5 mL) of S-alanine (1.5 g;
0.017 mol) was slowly added into stirring tetrahydrofuran
solution (200 mL) of succinic anhydride (1 g; 0.01 mol).
Resulting suspension was stirred at 30 8C for 1 h. Solid was
separated by filtration and filtrate was evaporated under
reduced pressure. Liquid residue was mixed with ethyl
acetate and the resulting mixture was dried over anhydrous
sodium sulfate. Drying reagent was removed by filtration
and filtrate was evaporated to solid residue and solid residue
was slurred in petroleum ether. Product was isolated by
filtration of white suspension in 79% yield. Mp 152–155 8C.
1H NMR (D2O, 500 MHz) d ppm: 4.35 (q, 1H, chiral CH;
JZ7 Hz), 2.69 (s, 4H, two CH2 groups of the succinic acid
moiety) and 1.42 (d, 3H, JZ7.5 Hz). 13C NMR (D2O, drop
of DMSO-d6 added for reference, 500 MHz) d ppm: 177.5,
177.4, 175.2 (carbonyl carbon), 49.4, 30.5, 29.8, 29.5 (four
aliphatic carbon). ES-MSK m/z 188.1 (85%, MKHC),
377.3 (100%, 2MK1HC), and 566.0 (30%, 3MKHC).
Anal. Calcd for C7H11NO5 (189.06): C, 44.45; H, 5.86; N,
7.40 Found: C, 44.37; H, 5.98; N, 7.33.
1
scopic). H NMR (DMSO-d6, 500 MHz) d ppm: 10.82 (s,
1H the pyrrole ring NH) 8.15 (d, 1H amide hydrogen, JZ
7.5 Hz), 7.52 (d, 1H benzene ring H, JZ7.5 Hz), 7.32 (d,
1H, JZ8 Hz), 7.14 (s, 1H the pyrrole ring CH) 7.05 (t, 1H,
JZ7 Hz) 6.97 (t, 1H, JZ7 Hz) 4.46 (ddd, 1H) 3.14 (dd, 1H,
J1Z14 Hz; J2Z5.5 Hz) 3.00 (dd, 1H, J1Z14 Hz; J2Z
8 Hz), 2.35 ppm (s, 4H aliphatic two CH2 group hydrogen).
13C NMR (DMSO-d6, 400 MHz) d ppm: 173.8, 173.5, 171.1
(carbonyls), 136.1, 127.3, 123.6, 121.0, 118.4, 118.2, 111.4,
109.9 (eight aromatic carbons), 53.1, 29.9, 29.1, and
27.2 ppm (four aliphatic carbon). MS-ESK m/z 303.1
(35%, MKHC), 431.3 (15%, 3MKCO2K2HC), and
607.1 (100%, 2MKHC). Anal. Calcd for C15H16N2O5
(304.30): C, 59.21; H, 5.30; N, 9.21 Found: C, 59.15; H,
5.38; N, 9.16.
3.1.12. Preparation of (R)-N-[1-carboxy-2-(1H-indol-3-
yl)ethyl]succinamic acid (2cR). The R stereoisomer was
prepared in 96% yield by following same procedure
outlined for the S stereoisomer. All spectra of R stereo-
isomer are identical to the S stereoisomer spectra.
3.1.8. Preparation of (R)-N-(1-carboxyethyl)succinamic
acid (2aR). The R isomer was prepared by following the
same synthetic procedure with 83% isolated yield. The
spectroscopic characteristics are identical to the S isomer.
Acknowledgements
We thank the Louisiana Board of Reagents for their
financial support (LEQSF (2001-04)-RD-B-12) for this
work.
3.1.9. Preparation of (S)-N-(1-carboxy-2-phenylethyl)-
succinamic acid (2bS). Tetrahydrofuran (500 mL)

926
B. S. Jursic, P. K. Patel / Tetrahedron 61 (2005) 919–926
Soc. 2001, 123, 7560–7563. (d) Guckian, K. M.; Schweitzer,
References and notes
B. A.; Ren, R. X.-F.; Sheils, C. J.; Tahmassebi, D. C.; Kool,
E. T. J. Am. Chem. Soc. 2000, 122, 2213–2222. (e) Ilhan, F.;
Gray, M.; Blanchette, K.; Rotello, V. M. Mocromolecules
1999, 32, 6159–6162.
1. For instance see (a) Chiral Drugs; Challener, C. A., Ed.;
Gower: New York, 2002. (b) The Impact of Stereochemistry on
Drug Development and Use; Aboul-Enein, H. Y., Wainer,
I. W., Eds.; Wiley: New York, 1997. (c) Rouhi, A. M. ‘Chiral
Business’, Chemical and Engineering News; 2003. May 5, pp
45–55. (d) Rouhi, A. M. ‘Chirality at Work’, Chemical and
Engineering News; 2003, May 5, pp 55–61.
7. Bentley, R. Arch. Biochem. Biophys. 2003, 414, 1–12.
8. For frontier orbital interactions see: Fleming, I. Frontier
Orbitals and Organic Chemical Reactions; Wiley: London,
1976.
2. For definition of stereochemistry see: Eliel, E. L.; Wilen, S. H.
Stereochemistry of Organic Compounds; Wiley: New York,
1994.
9. For similar enantiomeric recognition through hydrogen
bonding interactions in highly concentrated chloroform
solutions see: (a) Jursic, B. S.; Goldberg, S. I. J. Org. Chem.
1992, 57, 7172–7174. (b) Jursic, B. S.; Goldberg, S. I. J. Org.
Chem. 1992, 7270–7372. (c) Jursic, B. S.; Zdravkovski, Z.
J. Org. Chem. 1993, 58, 5245–5250. (d) Jursic, B. S.;
Zdravkovski, Z.; Zuanic, M. Tetrahedron: Asymmetry 1994,
5, 1711–1716.
3. For more information regarding cyclodextrin complexes see:
(a) Schneider, H.-J.; Hacket, F.; Ru¨diger, V. Chem. Rev. 1998,
98, 1755–1785. (b) Rekharsky, M. V.; Inoue, Y. Chem. Rev.
1998, 98, 1875–1917. (c) Song, L.; Purdy, W. C. Anal. Chem.
1992, 64, 1405–1412.
4. Chirality in Natural and Applied Sciences; Lough, W. J.,
Wainer, I. W., Eds.; CRC: Boca Raton, FL, 2002.
10. For p–p aromatic stacking in enantiomeric recognition see:
Jursic, B. S. J. Chem. Soc. Perkin Trans. 2 1994, 961–969. For
enantiomeric recognition in molecular aggregates such as
micelles see: Jursic, B. S. Tetrahedron Lett. 1993, 34,
963–966.
5. (a) Boeyens, J. C. A. Intermolecular bonding In Inermolecular
Interactions; Plenum: New York, NY, 1998. (b) Soltis, R. D.;
Hasz, D. Studies on the nature of intermolecular bonding in
antigen–antibody complexes Immunology 1982, 46, 175–181.
6. (a) Guan, L.; Hu, Y.; Kaback, H. R. Biochemistry 2003, 42,
1377–1382. (b) Lawrence, A. J. Chem. Rev. 1954, 54,
713–776. (c) Cubberley, M. S.; Iverson, B. L. J. Am. Chem.
11. Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P.
J. Am. Chem. Soc. 1985, 107, 3902–3909.
12. Connors, K. A. Binding Constants; Wiley: New York, 1987.